Maynard Handley (name99.delete@this.name99.org) on August 26, 2014 2:49 pm wrote:
> Klimax (danklima.delete@this.gmail.com) on August 25, 2014 10:12 pm wrote:
> > juanrga (nospam.delete@this.juanrga.com) on August 25, 2014 3:11 pm wrote:
> > > Klimax (danklima.delete@this.gmail.com) on August 25, 2014 2:14 pm wrote:
> > > > juanrga (nospam.delete@this.juanrga.com) on August 25, 2014 12:29 pm wrote:
> > >
> > > > > And TSMC has accelerated roadmap and will start 16nm volume production in 1Q15:
> > > > >
> > > > > Taiwan Semiconductor Manufacturing Company (TSMC) will advance volume production on its 16nm
> > > > > process to the first quarter of 2015 with monthly output of 50,000 wafers in order to meet demand
> > > > > for Apple's A9 processors, the Chinese-language Economic Daily News (EDN) has reported.
> > > > >
> > > > > TSMC originally planned to kicked off 16nm volume production in second-quarter 2015.
> > > > > TSMC faces strong competition from Samsung Electronics' foundry business.
> > > > >
> > > > > http://www.digitimes.com/news/a20140825PB201.html
> > > > >
> > > > > Broadwell-EP @14nm vs ARM server-class @16nm will be an interesting figth to watch.
> > > >
> > > > Only when against all odds TSMC actually delivers. So far their track record is supremely abysmal. So far
> > > > there is high probability ARM server chips on that node will go against later Intel chips, not Broadwell.
> > > > Also it seems that Intel fitted into their 14nm more changes then so far foundries to their "16nm".
> > >
> > > Broadwell-EP is scheduled for Q3 2015. Moreover, Skylake will use same node.
> > >
> > > Intel 14nm will be only about half-node ahead of TSMC 16nm. The gap has reduced.
> > They will likely e against that Skylake.
> > Half-node? They are not yet even fully on "20nm". (one baseband
> > modem and only limited availability) And it doesn't
> > look like they have even similar advantages on 16nm as Intel gets from 14nm. (Under discussion here)
>
> There is every indication that
> - the iPhone 6 will be announced on Sept 9
> - that its SOC is fabbed on TSMC 20nm
> - that it will be available for purchase sometime before the end of Sept
> - that 70 million or so will ship this quarter
>
> By all means argue about the reality of TSMC's 16nm node. But mocking the 20nm node at this date,
> and with these (highly relevant) facts comes across as just silly. To suggest that this 20nm
> product is "less real" than the more-or-less equally mythical at this stage Broadwell-Y (which
> doesn't appear to be in any product I can buy today) is the sort of hyper-legal quibbling that
> suggests one is more interested in scoring points than in understanding the situation.

Is it already announced and shipping and you can buy it? No. Wait until it is. Rumors are nice and dandy, but not really relevant, without say lost prototype. They are not yet facts. Don't get too far ahead.

> As for 16nm, Mike Muller (ARM CTO)'s keynote at HotChips 2014 gives some scaling
> results for 16nm along with a date of 2015. Interpret that as you like.

Well, as I understood David Kanter (and some other folks here, 16nm is just trigate on 20nm, so it doesn't look too similar to 14nm by Intel.